KR20230116675A - How to present points using point similarity and travel time - Google Patents

Abstract

An aspect of the present invention uses the points' similarity and travel time to determine the representation of the points. Certain aspects are methods applied to real estate. The method receives the destinations of the commute. For ease of use, this method determines properties with similar characteristics and similar commuting times within a given property cluster. Display of at least one property is presented to the user.

Description

How to present points using point similarity and travel time

[002] An aspect relates to presenting a search or comparison result using the similarity of points in one transportation system and the required travel time.

[003] When presenting the results of a search request to a user, search engines typically organize the results in order to reduce overprovision of information to the user and increase relevance. This organization typically uses two techniques: clustering and scoring. The purpose of clustering is to group similar search results in order to avoid providing repetitive information to the user. The purpose of scoring is to organize the search results in order to limit and present the most useful information to the user.

[004] Similarity means a case where each item is intuitively sufficiently similar. We use the term similarity in a broad sense consistent with the interpretation of the term by those with expertise in the field. Formally, the similarity can be typed from 0 to 10 by a mathematical function for each pair of items. 0 means the two items are not similar and 1 means the two items are similar. The number in between indicates the degree of similarity or dissimilarity. In some embodiments, two items are defined as similar if this number is above at least a threshold, for example 0.9. In the present invention, any item is considered similar by itself. Similarity is defined in various ways depending on the specific situation. In some embodiments, in text: two webpages are defined as similar if, for example, at least 90% of the text of the two webpages match, eg n=5 in n-grams. Other embodiments use numbers: for example, two listings of properties are defined as similar if their prices differ by less than 5% and are located in the same area. In another aspect, similarity is defined using artificial intelligence software in a computer system running on items. Examples of types of AI software include: Artificial Neural Networks, Support Vector Machines, Markov Models, Bayesian Networks, etc. For example, the similarity of two real estate listings is defined using artificial intelligence software running on images associated with these listings, and the similarity ranges from 0 to 1. In some aspects, similarity is defined for standardized items. For example, the text "San Francisco Cali." contained in one item is converted to "San Francisco, CA". Areas in square feet are converted to areas in square meters, and the color pixels in the image are readjusted to have an average brightness of 50%. In another aspect, similarity is defined using items represented by mathematical vectors or distances between vectors. For example, Chebyshev distance, Minkowski distance, etc. In another aspect, similarity is defined as: cosine similarity, string similarity (eg, Levenstein distance), semantic similarity, and the like. In some embodiments, the vector's coordinates are normalized, for example, to have a mean of 0 and a variance of 1. In some aspects, similarity is defined as combining at least two similarities. For example, after matching the text contained in the item using artificial intelligence to the image contained in the item, the two results are combined using, for example, a weighted sum. In another aspect, similarity uses only a subset of items. For example, ignoring information about mortgages in real estate listings. Without departing from the scope and spirit of the embodiments, many other ways of defining similarity will be apparent to those skilled in the art.

[005] There have been many studies on the problem of clustering. Examples of academic research include Jain, Murty and Flynn's previous study: “Data Clustering: A Review”, September 1999, ACM Computing Surveys, Volume 31, No. 3. Slightly oversimplified, given the number of items and the concept of similarity between items, the goal is to group items into similar items. Many clustering methods have been developed in previous studies. For example: linkage-based clustering such as hierarchical cohesive clustering; centroid-based clustering such as k-means clustering; variance-based clustering, such as an expectation-maximization algorithm; density-based clustering such as DBSCAN; Lattice-based clustering such as STING or CLIQUE; lexical clustering such as canopy clustering; These include subspace clustering such as CLIQUE or SUBCLU, and planned clustering such as PreDeCon. Some aspects of the clustering method compute clusters that meet additional requirements. Examples of requirements include: minimum or maximum cluster size, minimum or maximum cumulative similarity within a cluster, etc. Additional requirements are determined based on the context in which clustering is used.

[006] A common clustering method includes a method of calculating the similarity of each pair of items and classifying them into the same group when the similarity exceeds a critical point. However, when the number of items is large, the quadratic number of pairs of items makes this plain method impractical. To solve the scalability problem of second-order pairs, search engines often use self-learning methods to sort out pairs that are unlikely to be similar. An example of such a self-learning method is the previous study US 6658423 B1. In this sense, each item is a web page. In a heuristic method, a hash value is assigned to each web page. A hash value can be thought of as a very short text derived from a very long text on a web page. Web pages can be grouped based on their hash value (easily possible by sorting or bucketing hash values), and similarity is calculated only between two web pages with the same hash value. Since the hash value is obtained in that way, a pair of hash values of two web pages is designed to be equivalent to two web pages that are similar to each other. This is made possible using n-grams, for example. As a result, compared to the ordinary quadratic method, the self-learning method can greatly reduce the similarity to be calculated.

[007] Several self-learning methods have been developed to perform practical clustering in specific application areas. For example, if the items are real estate, the following self-learning methods are studied: US 20150012335 A1, US 9858628 B2 and US 10776888 B1. In addition, there are previous studies on self-learning methods when the items are job postings: US 10043157 B2 and Burk, Javed and Balaji: “Apollo: Quasi-duplicate Detection of Job Advertisements in Online Recruitment Area”, 2017 Data Excavation Workshop International Conference.

[008] A number of scoring methods have been developed. For example, US 7058628 B1, a study case scoring based on PageRank in a web search engine in a specific field, can be referred to, and also a study scoring based on characteristics of real estate in a specific field of real estate Example US 7974930 B2 can be cited.

[009] With the recent development of navigation technology, a search engine for searching and comparing properties of real estate using commuting time has been developed - see Example WO 2019164727. For example, the technology can quickly determine exact travel times from a large city to any property and place of work, given a user's work address. This enables in-depth searches in the real estate market. However, past methods may have fallen short of the goal of presenting these search results in a more useful way. Such presentation should address the problem of improving the relevance of repetitive information and search results in a pragmatic and extensible manner. The present disclosure presents a method for achieving this goal.

[010] A simplified summary of the present invention is presented in order to provide the reader with an insight into certain aspects of our claimed subject matter. This summary is not intended to be a general overview or exhaustive description of the scope of the invention, nor is it intended to identify key or critical elements of the invention. The purpose of this summary is to present some concepts in a way that is easy to read for someone with expertise in the field. Readers should refer to the disclosure of invention for details.

[011] Aspects of the invention include the following methods.

1. A method for determining the marks of a plurality of points included in a transportation system using similarity and length of travel;

Features of the method include:

(a) receiving a request containing at least one location included in the transportation system;

(b) determining at least two isochronous points included in the plurality of points;

where the length of travel in the transport system between each isochronous point and the at least one location is included in the range;

(c) Determining the indication using steps including one of the following:

i. determining a plurality of similar points included in the at least two isochronous points, followed by determining the representation of the plurality of similar points; or

ii. selecting at least one first point dissimilar to the at least one second point, both of which are included in the at least two isochronous points, then selecting the at least one first point and the at least one first point. determining the indication of two points; also

(d) Responding to the request with the indication.

2. A method of determining an overview of a plurality of points included in a transport system using the number and length of travel;

Features of the method include:

(a) receiving a request containing at least one location included in the transportation system;

(b) Calculating an array of two or more points included in the plurality of points, wherein,

i. At a first point and a second point included in the arrangement, the length of travel in the transport system between the first point and the at least one location is equal to the distance between the second point and the at least one location in the transport system. at least a certain range away from the length of the travel, and

ii. If the length of travel in the transportation system between a fourth point and the at least one location is around the length of the travel in the transportation system between a third point and the at least one location, then the The number associated with the third point is at most the number associated with the fourth point included in the plurality of points;

(c) determining the overview including an indication of the arrangement; also

(d) Responding to the request with the overview.

3. A method for determining an indication of at least two alternatives included in a plurality of objects of interest included in the transportation system;

Features of the method include:

(a) receiving a request including a point included in the transportation system;

(b) determining the at least two alternatives;

The length of travel in the transportation system between each alternative and the point is at most the smallest distance plus the threshold;

(c) determining the indication for the at least two alternatives to be non-unitary and not descriptive of travel; also

(d) Responding to the request with the indication.

4. a method for determining the indication of at least two points included in the transport system using the length of travel and the length of expected travel;

Features of the method include:

(a) receiving a request containing at least one location included in the transportation system;

(b) determining the lengths of at least two expected travels;

wherein the at least two expected lengths of travel include a length of expected travel in the transportation system between the at least one location and each point included in the at least two points;

(c) Selecting one or more points included in the at least two points using the lengths of the at least two expected travels;

Here, the number of said one or more points is at most a predetermined limit;

(d) determining the length of at least one travel;

wherein the length of the at least one travel includes the length of the travel in the transportation system between the at least one location and each point included in the one or more points;

(e) determining the indication of the one or more points using the length of the at least one travel; also

(f) Responding to the request with the indication.

[012] Aspects of the invention also include computer systems and tools that realize the above methods.

[013] The aspects of the invention presented in this disclosure are for illustrative purposes; These are not all. Many modifications and variations will be apparent to those skilled in the art without departing from the scope or spirit of the embodiments.

[014] In this presentation, words such as “first”, “second”, “above” or similar words are not used in any limiting sense, and are used for the purpose of distinction when not clear from the context. The singular includes the plural, except where the context clearly indicates. “Having,” “comprising,” “comprising,” or similar words indicate that there are elements or features, but do not exclude that other elements or features may be present or added.

[015] The drawings included in the disclosure illustrate various features and advantages of aspects of the invention:
* Figure 1: Shows an example of data flow within a real estate search and comparison engine using commuting time.
* Figure 2: Shows an example of a cluster of real estate listings included in an isochronous line.
* Figure 3 shows an example of a user interface that receives a request from a user and responds to the user.
* Figure 4: Shows an overview example included in the user interface.
* Figure 5: Shows an example of the steps for determining the presentation of different alternative theories included in the user interface.
* Figure 6: Shows an example of the steps of a two-step approach to determining travel time.
[016] The drawings are for illustrative purposes only. The other figures illustrate the invention within its scope and spirit as readily understood by those skilled in the art.

[017] The present invention relates to the general case of presenting points using similarity and travel time. However, for ease of explanation, we first illustrate an aspect of the invention in terms of a real estate search and comparison engine using travel time. For brevity, we call this the engine. This example is not limiting. The second half explains how this method is applied in the general case.

1 Exemplary Embodiment

[018] We describe exemplary aspects of the invention. In our description, we use the word module. In this field, the term module refers to a computer system that provides some specific function (and thus can be seen as a computer subsystem). In some aspects, this engine is divided into three modules: (1) aggregation module, (2) indexing module, and (3) submodules. Our choice of dividing the engine into specific modules is illustrative and not mandatory. Those skilled in the art will appreciate that the engine may be divided into different modules within the scope and spirit of the aspects.

[019] Next, the function of the module will be described. Throughout the entire description, we refer to Figures 1, 2, 3, and 4 and 5, where an apostrophe denotes at least one instance of a citing object. We describe this function as a sequence of steps in a specific order. However, this order is not limiting. Those skilled in the art will appreciate that certain steps may be performed in a different order, partially concurrently, or combined or omitted, within the scope and spirit of the embodiments.

1.1 Collection

[020] The collection module 1002 secures information on a real estate list from at least one information source (1001). These sources can be classified including, but not limited to the following types: (1) Directly responds to real estate companies (intermediaries), homeowners, construction companies, etc. injecting properties of real estate listings into the collection module. . or (2) indirectly, responding to websites that publish real estate listings, smartphone apps, etc., which are aggregated by the collection module. The first type of collection is carried out through a traditional call center, website or smartphone app, which provides a user interface for users to enter property characteristics. The second type of data collection is carried out by computer systems that use the Internet to retrieve data, such as websites or smartphone apps that disclose their own characteristics. Features are collected during retrieval. This process is often referred to as crawling. Characteristics of a property listing include, but are not limited to, at least one of the following: name, address, geographic location, when the property was entered or removed from the market, floor, total number of floors in the building, orientation of windows, view, Price, rent, deposit, information on lessor, information on loan, information on collateral, monthly management fee, contents of monthly management fee, moving date, land area, price per unit area such as 1 square meter, number of rooms, number of toilets, structure or layout, air conditioning, elevator situation, parking lot situation, swimming pool situation, garden or backyard situation, children's playground situation, gym situation, property written by landlord or agent, interior or view, indoor video tour, sound outside an open window Recordings of noise, measurements of air quality, status of brokers, ratings of brokers, identification tables separated by sources of information among real estate listings. In some aspects, the aggregation module obtains information of interest related to real estate listings through travel included within the transportation system. Any object of interest is an arbitrary place. Examples of subjects of interest and information about them include, but are not limited to: school (examples of information: type of school (public, private, daycare, elementary school, middle school, high school, university,…), school ranking, tuition fee ; restaurants, shops, convenience stores, laundromats, banks, ATMs, government offices, crime rates, police stations or military bases) In another aspect, the process of collection can be interpreted as measuring the real world, through which measurements the physical establishments described above Acquiring information about a physical presence may be in the form of data about an entity stored on a persistent storage medium by a source of information Within the scope and spirit of the embodiments, many other methods of obtaining real estate listings are within the scope of this field. It will be obvious to experts.

[021] In some aspects, the aggregation module operates continuously to create a state of the real estate market that changes over time. At some point along the way, there may be more than one list associated with any one real house, as they are created from multiple sources. Such a plurality of pieces of information may bring conflicting information, for example, when information is entered in different areas due to the misunderstanding of two brokerages. Listings produced at the same time can be modified later, for example, when the landlord corrects information about the property or uploads a new image of the property. Listings may be deleted or restored, for example, if a potential tenant cancels the lease. Listings may be intentionally ambiguous, such as when a landlord provides a range of real estate locations on the list, such as a 100-meter radius, or a range of floors in a building, such as “high rise,” to hide personal information. is the case There may also be cases where the listing is out of date, such as when a property has already been sold but has not yet been reflected in any source. There can be fake listings to fool potential renters, or malicious listings that deliberately distort features.

1.2 Index

[022] The index module 1003 standardizes the data acquired by the collection module into a format stored in a persistent storage medium. In some aspects, this format represents properties of real estate as feature vectors, with each feature associated with a value. Data types of values include, but are not limited to: text, number, image, movie, sound, and smell. Values are entered in a form accessible to the computer. For example, a range of values is represented by two characteristics: a number representing a high or low price. In some embodiments, non-numeric values are replaced with numeric values. For example, an elevator is classified as a feature of two values of “existence” and “absence”, and the values are applied from 1 to 0, respectively. In some aspects, the resulting features and numerical values are added to feature vectors. In some embodiments, a new feature or numeric value is added to a non-numeric feature of the feature. In some embodiments, a feature or value is calculated using at least one feature or value. For example, a unit price is calculated by dividing the value of a price by the value of an area, or by applying artificial intelligence software in a computer system to features or values. In some aspects, a feature has a value of a timetable indicating when a specific real estate listing was obtained by the aggregation module. In some embodiments, feature vectors are sparse, in the sense that certain features are missing from a real estate listing, but are presented for other real estate listings. For example, not all real estate listings have images of bathrooms. In some aspects, feature vectors are represented as hashmaps or lists. In some aspects, the indexing module reconciles conflicting information, for example by majority voting or averaging. To those skilled in the art, within the scope and spirit of the embodiments, many other ways of normalizing data will be apparent.

[023] Standardization is performed by a module tailored to a specific information source. This module is created by researchers in order to properly interpret the meaning of data obtained from sources. However, once written, this module runs automatically without human control.

[024] In some aspects, the indexing module creates one or more inverted indexes 1005 for at least real estate listings. Later, during request processing, a precomputed inverted index allows a quick search of a list of properties where a specific property of a property has a particular value. For example, an inverted index quickly finds listings of exactly three-bedroom properties. In some aspects, inverted indexes work with hash maps, filtered lists, etc., among others. In some aspects, the indexing module creates an inverted index for certain features, eg, each feature that commonly occurs in the course of a user request, eg: price, number of rooms, number of bathrooms, area, etc. Within the scope and spirit of this embodiment, many other methods for creating at least one inverted index will be apparent to those skilled in the art.

[025] In some aspects, the index model creates at least one clustering 1006 for a listing of real estate. Precomputed clustering is useful because it helps identify similar listings during request processing. Clustering uses a concept related to similarity, such as one of the concepts of similarity mentioned in the present disclosure. In some aspects, geographic distance is used by the module to create clustering of real estate listings. For example, a module reads a feature vector associated with a real estate listing from persistent storage. The geographic locations represented in this feature vector are clustered using a greedy approach. For example, if a place cannot be added to a previously created cluster without departing from the cluster radius, eg 10 meters, the places are processed in random order, and a place creates an independent cluster. Real estate listings are then assigned to clusters based on clusters of geographic locations. In some aspects, the indexing module creates clusters using the similarity defined in certain features, for example each feature that occurs frequently in user requests. In some aspects, the indexing module creates clusters using similarity defined by one or more features. For example, when creating only listings that are both considered similar in price and location. Other methods of creating at least one cluster, within the scope and spirit of the embodiments, will be apparent to those skilled in the art.

[026] In some aspects, the inverse indexing or clustering is limited by the aggregation module to lists obtained within a certain period of time, such as the last 24 hours. In another embodiment, the interval of this period is wide, for example 5, 6 years, etc., to show the trend of change in value over time. In another embodiment, the inverse index or clustering is limited to listings corresponding to properties on the market.

[027] In some aspects, the indexing module operates on interest in a manner analogous to the manner described above for operating on real estate listings.

[028] In some embodiments, the standardized data, inverse index or clustering, is stored on a persistent storage medium.

[029] In some aspects, the indexing module runs continuously, maintaining a current state of standardized data, inverted indexes and clusterings that change over time.

1.3 sub

[030] The submodule receives a request from a user, generates an indicator, and responds to the request according to the indicator. We describe several aspects below.

1.3.1 Indicators

[031] The submodule 1008 receives a request 1007 from a user. The request includes specification of at least one commuting destination (2001) (3002). A commuting destination is an arbitrary place, such as: a geographic place of interest, an address or a point. The request also includes parameters that determine how the at least one commuter destination forms the at least one commute route. A commuter route is a continuum of travel between the two end places. Commuter routes start or end at the property and include commuter destinations. A commuter route can form one of several forms, eg: a round-trip commute (e.g., home to work, back home, or home, to nearest school, back home), open-way commute Routes (e.g., from home, to school with a home included in the district, to piano academy, back home) to work, to piano academy, back home) or disconnected commuter routes (e.g., home to work, another home) ), or the travel route to the nearest school. Other methods of forming a commuter route, within the scope and spirit of the embodiments, will be apparent to those skilled in the art. Other parameters will be included in the request, eg: departure time, arrival time limit, probability of arriving within the time limit, means of travel (car, bus, metro, walking, combined, etc.) or travel frequency. For example, you can go to work 5 times a week and go to school 3 times a week. Other parameters may include parameters that include filtering limits that limit travel. For example: what type of car to use for the trip; maximum number of transfers; type of transfer permitted (eg, subway-bus transfer, bus-subway transfer); transit time; Limitations on walking time, or limits on which subjects of interest can be included in commuting routes (eg, only schools in the top 10). Given a particular property, we can determine the travel time for at least one commute route. We will present the results concisely as: Travel time between the property and at least one commuting destination. In some embodiments, a travel time between a property and at least one travel destination is a numerical value that reflects the total amount of time all occupants of the property spend for travel over a period of time, eg daily, weekly or monthly. can be These residents could be family members living on the property, roommates, co-workers, etc. Other parameters may include financial information about the user. For example: financial asset statement, credit rating, information about income such as hourly wages, job classification, salary, mortgage application; or information about the user's school, such as: admissions questionnaires, math test scores. The request may include specifications regarding desired characteristics or values of the property 3001, such as 80 to 90 square meters of area, desired characteristics, values of interest, and the like. Within the scope and spirit of the embodiments, other types of requests will be apparent to those skilled in the art.

[032] The submodule then finds the real estate listing L (2002) corresponding to the parameters of the request. For example, if the specification of the desired feature is an area between 80 and 90 square meters, the module finds all listings within the range between 80 and 90 square meters. In some embodiments, the search process uses an inverted index generated by an indexing module. For example, by intersecting a list search set associated with the desired characteristics and their values specified in the request. In some embodiments, the search is limited to listings that the aggregation module has acquired, for example within the last 24 hours. In another aspect, the search is limited to real estate on the market. In another embodiment, L contains all real estate listings. Within the scope and spirit of the embodiments, other methods of finding real estate listings that match the property sought will be apparent to those skilled in the art.

[033] Then, the submodule takes at least one or more travels between each geographic location in list L, using any method known to experts in the field, such as the navigation service 1010 of previous studies mentioned in the disclosure. Determine a time and at least one commuting destination. For example, for a specific geographical location H, a specific geographical location of work, and a specific geographical location of school among real estate listings, the navigation service determines the travel time D _w from H to work and back to H, and from H to school, Calculate the required travel time D _s to return to H again. Then, the required travel time between the real estate listing and at least one commuting destination is D _w + D _s , in which D _w contribution and D _s contribution are included. In another aspect, the travel time required is derived from another travel time required. For example, the required travel time becomes a (5D _w + 3D _s ) / (5 + 3) weighted sum value according to the travel frequency, for example. In other cases, the travel time is the difference between the travel times, eg, the travel time to the current home (in the request) minus the travel time to the new home candidate in list L. will be. In some embodiments, the travel time is the shortest travel time. For example, it is the time calculated using Dijkstra's algorithm for a graph representing a transportation system, and this graph was created using a previous research method as mentioned in the disclosure. In some embodiments, the expected travel time is, for example, a factor of 2 plus a factor of 15 minutes, or an expected travel time that falls within a multiplier factor or within an additional factor of the shortest travel time of 1,000 meters (the characteristics of such expected travel time are described in the disclosure). Corresponds to the travel time using the previous research mentioned in the method or disclosure). In some aspects, the navigation determines the travel time using methods from previous studies such as:

a) Representatives are places that occur frequently in the shortest travel, such as endpoints on both sides. Like WO 2019164727, representative sites have the following characteristics: are places included in a transportation system, where many representative sites are the size of the transportation system, no matter how large, multiplied by a predetermined ratio of at most 1, examples of representative sites include do:

(i) Landmarks, Portals, Hubs, Beacons, Seeds, Transportation Junctions, among other studies, by Sommer: “Shortest-Path Exploration in Static Networks” ACM Computing Surveys, 2014, Vol. 46 (4),

(ii) transit stations or global stations where transfers may occur during long connections, such as US 8756014 B2;

(iii) the center of a grid as in CN 105975627 A;

(iv) boundary vertices as in US 9222791 B2;

(b) processed graph data including connection points representing a pre-filtered map, such as US 9250075 B2;

(c) reduced road maps such as US 9195953 B2;

(d) a hierarchy of polygonal layers, such as US 7953548 B2;

(e) a grid as in CN 105975627 A;

(f) a subgraph obtained by excluding at least one intermediate point, as in US 8949028 B1;

(g) one or more anchored junctions, such as EP 2757504 A1;

(h) a forward partial path and a backward partial path, as in EP 1939590 B1;

(i) Overlaid graphs such as US 9222791 B2

(j) midpoint as in US 20110251789 A1; or

(k) transit between a station of departure and a destination, i.e. at or near a place of departure and a station of destination, as in US 8417409 B2, US 8738286 B2, US 8756014 B2, us 10533865, KR 101692501 B1, or CN 104240163 A; at or near the target site;

Or any of the methods known in the art include:

(l) Geisberger, Sanders, Schulte and Delling's contraction layers: “Contraction Layers: Fast and Simple Paths in Road Networks” 2008 Workshop for Experimental and Efficient Algorithms; Shortest Batch Path” 2011 Workshop on Algorithmic Approaches for Transport Models, Optimizations and Systems;

(m) Techniques based on CRP, GRASP and PHAST by Baum, Bourhold, Divelt and Wagner: “Fast and Accurate Isocontour Computation in Road Networks”, Journal of ACM Experimental Algorithms, October 2019;

(n) Techniques described in Sommer's research: “Finding Shortest Paths in Static Networks,” ACM Computing Surveys, 2014, Vol. 46(4); or

(o) Bast, Delling, Goldberg, Müller-Hahnman, Pasor, Sanders, Wagner and Verneck: “Route Design in Transport Networks” Algorithm Engineering 2016.

In some embodiments, before finding L, the submodule determines the travel times for geographic locations in the lists, stores the travel times for the geographic locations in the list, stores the travel times to a computer readable persistent storage medium, and, after finding the places within L, determines the travel times from the storage medium. Call up the time required. In some aspects, a sub-module determines the travel times of sub-lists. For example, it means only the lists within the vicinity of a large city, such as a radius of 500 meters or 1 minute from a predetermined place. Within the scope and spirit of the embodiments, many other methods of determining travel time will be apparent to those skilled in the art.

[034] The serving module then groups the list L according to the proximity of the travel time. A group is related to the concept of an isochrone, which refers to a line connecting points with the same travel time on a map starting from a certain place. However, we use the concept of isochronous zone (2003) in a broader sense of range. In some embodiments, the range is set to a small distance, such as 500 meters. In some embodiments, a range includes at least two or more values. For example, a submodule considers a continuous range of M minutes: [0, M), [M, 2M), [2M, 3M), etc., for example, where M is set to 15. Group i is associated with a range of [iM, (i+1)M). Then, if D is the travel time between a geographic location in the list and at least one commuting destination, the list is defined as a group i, the range associated with this group including D. Another example is that a group contains a limited number of listings, such as 1000 listings in each group, that appear consecutively in a sequence of listings organized by travel time. In this case, the range may have different widths. In some embodiments, a group may include a list of real estate without any geographic restrictions, such as 1 hour east or 1 hour west of work (2004). In some embodiments, a group may include real estate listings 2005 limited to areas around the metropolitan area, such as a 500-meter radius or one-minute travel from a pre-determined location. In some aspects, ranges may overlap. In some embodiments, a group corresponding to an isochronous zone is computed that includes a range of travel times, such as a predetermined range or a range specified in a request. Within the scope and spirit of the embodiments, many other methods of using proximity of travel times to create group listings will be apparent to those skilled in the art.

[035] Subsequently, the submodule determines clusters of a similar list in any group (2004) (2005). In some aspects, the sub-modules use any concept of similarity mentioned in the disclosure. In some aspects, the sub-modules use the concept of similarity discussed in the present disclosure. In some embodiments, clustering is determined using the clustering methods described in the present disclosure. A cluster may have only one list if there are no other similar listings, and a cluster may contain a plurality of listings. In some aspects similarity is influenced by the user. For example, a request could be: "Ignore multiple brokerages", in which case listings advertising similar homes from different brokerages would be considered similar. In some aspects, pre-computed clustering is used to expedite the clustering process during request processing, e.g., lists are pre-clustered using textual features using pre-computed clustering, each dictionary - Clusters are clustered by user-influenced similarity. Within the scope and spirit of the embodiments, many other methods of determining clustering for a group will be apparent to those skilled in the art.

[036] The submodule determines the score of lists or clusters of lists. One score is an entity that is compared to another score to establish a ranking. For example, some scores are numerical values compared using “greater than”. In some embodiments, a score in a list uses at least one feature and its value, examples of which include: characteristics of travel (eg, travel time between a geographic location and at least one commuting destination); Characteristics of interest (e.g., ranking of schools; characteristics and values of interest included in at least one commuting destination; characteristics of location (e.g., crime rate; availability of schools or local services and amenities; etc.) , monetary characteristics (such as standard market price, price increases or reductions), reputational characteristics (such as previous leases brokered by realtors, previous buyer reactions, etc.), and temporal characteristics (such as how long the listing has been on the market). In one aspect, the score is defined as the “features” listed in the previous study US 7974930 B2. In one aspect, the score is calculated by artificial intelligence software executed by a computer system, which acts as a basis for predicting the score. , web links associated with the feature, values and requests, and past interactions between the user and the submodule that have been mastered. In some embodiments, a score is determined using an arbitrary mathematical formula. is the weighted sum of the numerical features. Weights can be negative, positive or 0. In other embodiments, we use the same weights. In some embodiments, each feature can affect the score Weights are determined so that they have equal opportunity For example, features with high median values in all lists are given a lower value In some embodiments, weights are used to emphasize specific features, for example monetary features use a higher weight for lists that are more recent, and higher weights for more recent listings. If , the score of that list will go up, eg by adjusting the weights appropriately. In some aspects, the cluster score of a list can be a mathematical statistic of the scores of lists within a cluster. For example: the maximum score for any list in a cluster, or the weighted sum of numerical feature values in a cluster. In some embodiments, a score is a vector whose coordinates are computed using at least one feature and its value. For example: For a 2-dimensional vector, the first coordinate is the negated travel time and the second coordinate is calculated using the features and their values as above. In some aspects, these vectors are organized in a lexicographical fashion. In some embodiments, scores are textual, such as the names of apartment complexes arranged alphabetically. Within the scope and spirit of the embodiments, other methods of determining the score will be apparent to those skilled in the art.

[037] The submodule selects clusters. In some aspects, the module selects a number of the highest scoring clusters, eg, the top 20 clusters. In some aspects, a module selects multiple clusters that satisfy at least one additional requirement. In some aspects, an additional requirement is to select clusters within a range of travel times, for example in the range [0, M]. In some embodiments, an additional requirement is to select at most a certain number of clusters, such as, for example, 5 clusters. In some aspects, an additional requirement is the geographic sparsity of the selected clusters, eg, selecting at most a predetermined number of clusters in a neighborhood. For example, the clusters are processed in order of score, with the highest score first, and if two or more clusters above the threshold are selected in the neighborhood, for example, in the length of the travel within 500 meters or 3 minutes, This is to ensure that no other clusters are selected. In some aspects, an additional requirement is that the selected clusters have different feature values. For example, choosing a two- or three-bedroom apartment. In some embodiments, diversity is secured by a greedy process and excluding selection of clusters having characteristics similar to feature values of previously selected clusters. In some aspects, additional requirements are specified in the request. For example, a request may ask for a list of specific school districts. In some aspects, the module selects the clusters using any of the clustering methods mentioned in the disclosure. Here, the clustering method acts on each item in the cluster. For example, a clustering method determines which item is the centroid, and this item becomes the cluster. Within the scope and spirit of the embodiments, many other methods of selecting clusters will be apparent to those skilled in the art.

[038] In some aspects, the submodule determines an indication of at least one or more listings in the cluster and responds 1009 to the request with the indication. For example, as described above, clustering is determined using similarities among lists and similarities between geographic locations on lists and travel times between at least one commuting destination. Aspects of the indication include, but are not limited to, at least one of the following:

(a) the list with the highest score within the cluster;

(b) position 3007 for a score in a list within a cluster;

(c) a list of travel times (3003) and (3004);

(d) part of the travel time, eg walking time included in travel by public transport;

(e) part of the list (3005); i.e. a literal representation of a list

(f) a web link 3006 to a list published by the source;

(g) at least one characteristic or value 3005 of a list;

(h) mathematical statistics of values, such as: (i) a histogram of the values of a feature, (ii) the value of the most or least occurring feature, (iii) a random sample of the values of the feature, (iv) the maximum or minimum of any numerical value; (v) the mean, median, percentage, standard deviation or variance of any numerical value;

(i) mathematical statistics based on time period, eg price trends over the past five years;

(j) a summary of any one list in the cluster;

(k) a summary of at least two listings in the cluster, e.g.: (i) the number of listings (3007) or (ii) an indication of the difference between the two listings, e.g. which listing has a lower price or is more reliable Indicates that it was posted by an intermediary.

(l) a combined listing obtained by combining the characteristics and values of at least two listings in a cluster, eg: giving the phone numbers of all brokers advertising a particular house, but only advertising the number of rooms in this house once;

(m) graphical or textual representation of clusters; or

(n) Consistent with the above, but relating to a subject of interest included in at least one commuting destination. In some aspects, for each of the selected clusters, selected clusters are determined including an indication of at least one list.

In some aspects, a representation of an arbitrary set of real estate listings is determined according to the method above in which representation of at least one listing in the cluster was determined. For example, it includes mathematical statistics about the price of listings currently on the market, or of listings that meet user requests. Other methods of determining the indication, within the scope and spirit of the embodiments, will be apparent to those skilled in the art.

[039] In some aspects, we expressly restrict the presentation of at least one list, such that the presentation does not display information other than what is explicitly stated in the restriction. These constraints include indications from at least one of the above lists, but add “only” validators, e.g.:

(a) only those fragments of a list that have the highest score in at least one list;

(b) only a combination of two marks, or

(c) Only combinations of marks k when k > 3.

This restriction has the advantage of increasing relevance by reducing the excess of information provided to the user. For example, presenting only one list among all lists between 10 and 20 minutes of travel time imposes only a small cognitive load on the user and provides only useful information to the user.

[040] In some aspects, the submodule responds 1009 to the user with only an indication of at least two dissimilar clusters. For example, this method selects two clusters where the distance between them is at least a critical point. Distances between clusters are well known in the art. For example: a pair of lists between two clusters or the minimum distance between the centroids of two lists. An example of a critical point is 1000 meters, or 1 minute of travel. The method then determines the indication between the two clusters, which helps to generate various responses to the request. These at least two clusters are determined, as in the case described above, using the similarity of the lists or the geographic location of the lists and the travel time between at least one commuting destination. Thus, the first list in one cluster can be considered dissimilar to the second list in another cluster. In one aspect, the required travel time is in a certain range, but in another aspect, the required travel time does not have to be in a certain range. Many other ways of responding to the indication of at least two clusters within the scope and spirit of the embodiments will be apparent to those skilled in the art.

1.3.2 Overview

[041] In some aspects, the method determines an overview of real estate listings using the amount and travel time associated with each listing. The beauty of this overview is its ability to help users find listings that result in a customized balance between quantity and travel time. We interpret the number as a discount in some aspects. This interpretation is not limiting. Other ways of interpreting the number within the scope and spirit of the embodiments will be apparent to those skilled in the art. To briefly explain how overviews are determined, we will refer to individual lists. However, these individual inventories should be understood to correspond to clusters of similar inventories determined according to the methods described in this disclosure. Each travel time referred to in this description is the time between a property and at least one commuting destination.

[042] In some aspects, the method computes a series of listings with short travel times and various travel times to obtain a low number of values relative to nearby travel times. In some embodiments, this calculation starts with a list L that matches the desired feature. In some embodiments, this calculation is performed in a greedy manner. For example:

(a) This method considers lists U with sale prices and arranges these lists in order of sale price, with the lowest price first;

(b) the method then processes the listings in order of sale price;

(c) During this process, this method selects the next list, ￡, and then removes the processed travel time from any list that is within the travel time range of ￡;

(d) The method then repeats step (c) until there are no more lists to process.

This calculation yields any number in the list where k > 1. This number k depends on how wide or narrow the range is. For example, this range is set to 15 minutes or 1,000 meters, etc. We want these lists to be £ _1, £ _2, £ _3, … Like ,￡ _{k ,} they are listed in increasing order according to the required time. For example: £ ₁ (4001) is a travel time of 5 minutes and a selling price of USD 1 million; £ ₂ (4002) has a travel time of 14 minutes and a selling price of USD 1.2 million; ￡ ₃ (4003) has a travel time of 25 minutes and is priced at USD 700,000; etc. Due to the way these listings are calculated, we can see that there are no listings whose travel duration is around the travel duration ￡ _i at a lower price. The neighborhood includes a range of travel times of the required travel time ￡ _i at the largest, or a range of the required travel times of the required travel time ￡ _i at the smallest. A neighborhood can include both ranges. The definition of this neighborhood is the extent to which the selling price does not change over a long period of time along the travel time. For example, in the example above, £ ₁ is a list of the cheapest properties with a travel time of around 5 minutes. In some aspects, we compute an array of listings with relaxed requirements for travel duration or amount. For example, we select, for example, the critical points of the list within the travel time range of at most 5 lists. For example, we select the smallest positive value as a threshold, e.g., at most, e.g., five lists. In some embodiments, the computed lists are arranged arbitrarily (i.e., £ ₁ , £ ₂ , £ ₃ , …, £ _k do not necessarily follow an increasing rate of travel time; e.g., a decreasing order). or the unchanging order of the quantities associated with the computed lists). Many other ways of computing an arrangement of lists within the scope and spirit of the embodiments will be apparent to those skilled in the art.

[043] In some embodiments, the method calculates the list £ ₁ , £ ₂ , £ ₃ , . . . ,￡ Determines the display of _k . The indications of the clusters are explained in the disclosure, and there are corresponding indications that are readily recognizable to experts in the field within the scope and spirit of the embodiment £ ₁ , £ ₂ , £ ₃ , . . . , ￡ _k . For example, the display has the sale price and some information for each computed listing. In some embodiments, this indication includes a mathematical statistic for the selling price of listings whose travel times are within a range from a travel time of ￡ _i . Exemplary mathematical statistics are described in the disclosure. For example, the 10th percent, the 50th percent, and the 90th percent. By doing this, it is possible for the user to guess how much the price is possible by comparing the travel time required with the travel time required for ￡ _i . In some embodiments, mathematical statistics are included for the overall price of List U. In this way, the user can guess what the selling price on the market is generally. For example, a user may consider a two-bedroom apartment currently on the market in the metropolitan area to be cheap (eg, 10 percent) (4004) or expensive (eg, 90 percent) (4005).

[044] Calculated list ￡ ₁ , ￡ ₂ , ￡ ₃ ,… , ￡ _k can be viewed as a “first order” list, which is an overview of the cheapest prices based on travel time. In some aspects, this method computes "secondary" lists. For i, the method considers ￡ _i and the following ￡ _i+1 and their associated travel times d _i and d _i+1 (4006). The method then considers U _i , a subset of list U whose travel time is between d _i and d _i+1 . Then, this method can determine the representation of U _i using any of the methods described in the disclosure. For example, the indication of U1 includes a small number of geographically dispersed listings with travel durations between 5 and 14 minutes and the cheapest listings with travel durations in this range, at most 10 or less. The representation of U _i may include mathematical statistics about U _i . The sets of limit-cases are defined by: U ₀ where U ₀ contains all lists in U for which travel times are less than d ₁ , and U _k contains all lists in U where travel times are less than d _k . . In some embodiments, the limit-case is further limited to within the travel time. The U _i set may be empty, eg if there is no preferred list of travel times that fit between d _i and d _i+1 . Therefore, “second priority” may not be constant. In some aspects, this method computes "secondary" lists by dividing the travel time into sub-units and computing an array. In some embodiments, the method continues by dividing the travel time into smaller segments to calculate a "third priority".

[045] In some aspects, the method determines an overview, which is a graphical representation of the relationship between selling price and travel time. In some aspects, the graphical representation depicts mathematical statistics about the selling price associated with each travel time span of travel. For each range of travel times, this method determines a mathematical statistic for the selling price and creates an encoded shape of this mathematical statistic. For example: dots, lines, ovals, rectangles, range bars, etc. Graphical representations take many forms, including but not limited to: plots, histograms, pie charts, and heat maps. For example, the plot includes: a horizontal axis corresponding to travel time, a vertical axis corresponding to selling price, and a mathematical statistic for the selling price associated with each listing of travel time. For example: (4010) represents the 90th percent of the sales price, (4011) represents the median of the sales price, and (4012) represents the 10th percent of the sales price. For example, Fig. 4 shows that among the list of ranges including 30 minutes of travel time, the sales price corresponding to the 10th percentage is US$ 1 million. Many other methods of determining a graphical representation of the relationship between selling price and travel time within the scope and spirit of the embodiments will be apparent to those skilled in the art.

[046] In some aspects, the method enables navigation. This method is performed by the device. This device shows an indication of “priority 1” listings. Users can interact with this device. For example, the user interacts with the user interface element 4007 (tapping on the display screen, operating a mouse, touching, rubbing, touching and holding if sensitive to touch, etc.), and thus In response, the device interacts with the user interface element 4008 to show or hide the display of appropriate "secondary" listings. As a result, the user is presented with a concise overview of the lowest price, with which the user can repeat searches to explore trade-offs between selling price and travel time. In some embodiments, if the presentation for lists is: linear (eg, a list), the user can scroll the presentation up and down; if circular (eg, the list is folded into a circle), the user can rotate the presentation. and can rotate in reverse. In some embodiments, the locations of the "second priority" listings appear on the map when the "second priority" listing appears. In some aspects, the locations of the other listings are hidden from the map when a "secondary" listing appears. In some aspects, a user interacts with the graphical representation. For example, when a user taps a user interface element 4013, in response, the device displays a list of lists whose travel times are within the range associated with this element. Within the scope and spirit of the embodiments, many other ways of finding one's way will be apparent to those skilled in the art.

[047] This method uses the selling price and the travel time to list ￡ ₁ , ￡ ₂ , ￡ ₃ ,… , use different methods to calculate ￡ _k . In some embodiments, the method utilizes any clustering method described in the disclosure. Each ￡ _i is selected from the cluster, for example, the centroid of the cluster. In some embodiments, at least one additional clustering requirement is set and includes: set k to a maximum value such as 20; ￡ if _k is the smallest possible; If the minimum range of the travel time ￡ _i and the travel time ￡ _i+1 is set to a minimum such as 10 minutes; If you set the selected list to the maximum value, such as 5, around the travel time required; In the case of clustering only lists of minimum travel times, such as the minimum time of 2 hours; to cluster only listings with the lowest prices, such as 75 percent of the lowest prices; When ￡ ₁ is selected from a list of constant travel times, such as within the range of 5 minutes, which is the minimum time; In case of selecting ￡ _i with the lowest price of any list within 10% of the lowest price in the range including the travel time of ￡ _i , etc.; In some aspects, the clustering problem of at least one additional requirement is coded into a linear program. In some aspects, ranges of travel times differ from other ranges of travel times. For example, travel times corresponding to relatively low-cost listings are narrower. In effect, the lists £ ₁ , £ ₂ , £ ₃ ,… , ￡ _k does not have to be the same interval of travel time. Within the scope and spirit of the embodiments, many other ways of computing lists will be apparent to those skilled in the art.

[048] Previously, we used the selling price as a specification of the number. Generally, a number is a body that can be compared with other numbers to establish an ordering. In some embodiments, the number is set to a feature value, such as, for example, a ranking of schools. In some embodiments, the request describes a number, eg if the request is “prefer high price” then the number is the negation of the sale price (i.e. the sale price is multiplied by -1). In some aspects, the number is derived from the value of the feature using a mathematical formula. For example, the number reflecting the "centrality of floors" of a particular list is calculated using the formula (f/b - 0.5) ² , where b is the number of floors in a building and f is the number of floors in a building in that list. is the number of floors In some aspects, it shows derived text associated with derived numbers such as "centrality of layers" rather than simply "several layers". For example, the number reflecting the geographical similarity is determined as the absolute value of the difference in direction along the shortest arc (eg, for a user request specified as a north window, the number is 90 degrees to the east window). In some embodiments, the number is derived from the values of two or more features, for example by dividing the selling price by the area. In one aspect, the number is determined from the expected features from the desired features. For example, when a user searches for an apartment of about 100 square meters, the count is |x-100| , where x is the square meter of the apartment. In some aspects, predictions are automated based on requests and characteristics, eg: set rules; For example, skilled artificial intelligence is used according to past requests. In some embodiments, a number equals a score. Within the scope and spirit of the embodiments, many other methods of determining the number will be apparent to those skilled in the art.

[049] In some embodiments, element 4002 may define an indication of at least one list A, element 4003 may define an indication of at least one list B, and both elements A and B are included in the isochronous zone E, where A is dissimilar to B. In some aspects, element 4009 may be defined as an indication of at least one list C, and also contained in the same isochronous zone E, where A is not similar to B and B is not similar to C. . In some aspects, the isochronous zone E is set broadly. For example, when there is a request for an apartment, the range is at least 160 minutes, when there is no apartment available between 2 and 180 minutes of travel time. In some aspects, the indication of the plurality of points includes an overview of the plurality of points.

[050] Many other methods of determining the overview of real estate listings using the number associated with each listing and travel time will be apparent to those skilled in the art.

1.3.3 Alternatives

[051] Consider the case of two schools with a school district containing a certain house. In this case, the child living in this house may attend either school. Once admitted, the child attends the chosen school and does not go to another school (the school not selected). Thus, the selected school contributes to a household's total commute time. Although other schools do not contribute, it is useful to know that this school can provide other alternatives, because other schools may contribute at other times.

[052] Thus, when we want to determine an alternative object of interest, a general problem arises of reducing the overload of information on the user and increasing its relevance. We present our solution to this problem.

[053] Let's start with the archetype for alternative schools. Our method works for school S ₀ (5003) and real estate H (5002). In one aspect, S ₀ and H are included in the request. And S ₀ is the school closest to H in terms of travel time or distance. In one aspect, S ₀ is included in the request, and H is determined by steps of our other method. In some embodiments, both H and S ₀ are determined by steps of our other method. We determine the travel time D ₀ (5001) between school S ₀ and real estate H using any of the methods described in the disclosure. Then we determine a set A of alternative schools whose travel time D ₀ is not much greater than school S ₀ . For this purpose, we select the nearest m > 0 alternative schools S ₁ , . . . , S _m (5004) (5005) (5006). We determine the travel time D _i between school S _i and property H. Then, for each i > 1, we add a critical point to D ₀ to evaluate whether D _i < t for a given travel time t (5007). Examples of critical points are: 20 minutes, 2000 meters, etc. If the evaluation succeeds, school S _i is included in set A. In some aspects, school S _i is removed from set A if the evaluation fails. As a result set A could be nothing, eg all alternative schools are very far from property H. In some aspects, we include school S ₀ in set A. In some aspects, school S ₀ is removed from set A. In some embodiments, set A includes at least two schools.

[054] Then we decide on the representation of set A. These indications include, but are not limited to:

(a) conforms to the indications defined in Section 1.3.1;

(b) the meaning of set A, i.e., text such as “average ranking of alternative schools”.

(c) information obtained by the collection module about the schools included in set A, for example: (i) name of school, (ii) type of school, (iii) ranking of school, (iv) tuition fee of school, (v) ) the school's criteria for admission or (vi) the likelihood of admission;

(d) mathematical statistics about the information obtained by the aggregation module about the schools included in set A, for example: (i) the number of schools, (ii) the highest, lowest or middle ranking of schools, (iii) the schools the highest, lowest or median cost of tuition for students, and (iv) their chances of getting into school, e.g., 1-PROD _{i in A} (1-p _i ) , where P _i is S _i school availability, (v) the expected value of the rank, e.g. SUM _{i in A} (p _i *r _i ) , where r _i is the school's ranking, or (vi) the expected value of tuition, such as SUM _{i in A} (p _i * u _i ) , where u _i is the tuition fee of the school.

(e) A mathematical statistic about the travel time for the schools in set A, such as: (i) the travel time D _{i ,} or (ii) the maximum, minimum or average commute time or

(f) A combination of the above, eg: (i) a weighted sum of the rankings of schools, e.g. weighted by D _i / (SUM _{j in A} D _j ) as a fraction of travel time D _i or (ii) commute time weighted sum of, i.e., admission probability P _i .

[055] In some aspects, the representation of set A includes a description of the travel. In some embodiments, an indication about set A includes a non-travel description , eg, the number of schools in set A. In some respects, the notion of “not a description of the travel” is more limited, in order to exclude previous research, which is clear to experts in the field.

[056] In some aspects, a representation of set A is a non-unionary dependency on set A. If set A is limited to at most one school, then this dependence is singular. Other dependencies are non-unitary. For example, in set A, the school name is singular. However, in a set A containing two or more schools, the name of the top school is non-unity. A non-unity representation is useful because it summarizes information about a large set A. One aspect of a non-unitary representation is a mathematical statistic, which is a function of the numerical values for each school in a set A containing at least two schools, the mathematical statistic having a non-zero partial derivative for each numerical value. In some embodiments, the average ranking of at least two schools is a non-single indication. In some aspects, the representation for set A includes a singular representation. In some aspects, the notion of "non-unitary", as will be clear to experts in the field, is more limited to exclude previous research.

[057] In some aspects, we explicitly constrain the representation of set A so that the representation contains no information other than those explicitly listed by the constraint. This restriction includes any of the above indications of set A, but adds a validator of "only", for example:

(a) only the names of the top ranked schools included in set A;

(b) only the number of schools;

(c) average commute time only

(d) the school's average ranking only

(e) only a combination of the two marks, or

(f) Only k combinations of indications when k > 3.

This has the advantage of increasing relevance by reducing the plethora of information that restrictions provide to users. For example, if only the number of schools in set A is presented as the only information on set A, the burden to be recognized by the user is reduced and useful information can be provided.

[058] In some embodiments, prior to determining the set A, alternative schools S ₁ , . . . S _m 's are filtered out. Some embodiments use a filtering method based on property H in the school district that contains the eligible school. In some aspects, filtering is on a request basis. For example, if a request is “Only private schools”, “Only within 20 minutes from home”, “Only tuition is less than $500”, “Only schools within the top 30%”, “School level that my child can pass Only”, “Only schools with at least an 80% chance of passing in light of our child’s characteristics… ” can be specific. In some embodiments, we use a predefined filtering method based on a mathematical formula using the travel time D _i or school characteristic values. Exemplary mathematical formulas represent the filtering methods specified in the request.

[059] In some embodiments, schools are scored in a manner similar to the method for scoring lists described above. Accordingly, the scores of the schools are used to select set A.

[060] In some aspects, a score for real estate listings is determined using information obtained in the aggregation module for the schools included in set A. For example, the score is increased by the mathematical formula value of set A.

[061] In some aspects, an indication of a plurality of points includes an indication of set A. For example, if the request specifies that properties be ordered by school order, then the property's representation includes an indication of the nearest school and an indication of the alternative schools in set A. In some aspects, a submodule responds to a request for display of set A.

[062] The above method of indicating alternative schools can be generalized to indicate alternative objects of interest. However, how do we get the alternative interest S ₁ , . . . , it is necessary to discuss whether to determine S _m . Whether two objects of interest can be considered alternatively depends on the specifics of the object of interest and therefore may be arbitrary. Thus, our method automatically determines alternatives in the same way as the views of experts in the field. For example, we find that if the alternative interest is a hospital and the request is specifically "orthopedics", the decision is based on hospitals with orthopedic units. In some aspects, we have an alternative interest S ₁ , . . . , S _m is determined using the similarity of the objects of interest.

[063] Within the scope and spirit of the embodiments, many other methods of determining alternative indications of interest will be apparent to those skilled in the art.

1.3.4 Two-step approach

[064] In some aspects, the sub-module uses a two-step approach: (step 1) computes the expected travel time and uses it to select some clusters. Subsequently (step 2), required travel times of the selected clusters are calculated, and a display is determined using the required travel times. In some aspects, the two-step approach allows the information source to be saved (step 1) and limits the quality of the travel from deteriorating (step 2). For the sake of brevity, we describe an aspect with only one commuting route (6001). However, it will be clear to experts in the field how to generalize this aspect to at least one commuter route within the scope and spirit of the aspect. This aspect is applicable to any set of lists and includes:

(a) Calculate an estimated travel time between each geographic location in a defined list and a commuting destination, aspects including:

(i) find a nearby representative 6003 within a threshold of a commuting destination 6001, find a nearby representative 6005 within a critical point of a geographic location 6007, pre-select between two nearby representatives Retrieves the calculated travel time 6004, followed by an estimated travel time to the retrieved travel time, optionally incremented by: travel time (6002), or travel time (6006) between a geographic location (6007) and a nearby representative location (6005);

(ii) find a nearby representative location 6003 within the threshold of the commuting destination 6001, retrieve the pre-computed travel time 6008 between the nearby representative location 6003 and the geographic location 6009; Subsequently, the estimated required travel time is determined in the imported travel required time, which is optionally increased according to the required travel time (6002) between the commuting destination (6001) and the nearby representative location (6003).

(iii) retrieve the pre-computed travel time 6010 between the commuting destination 6001 and the nearby representative location 6011, find the nearby representative location 6011 within the threshold of the geographic location, and The estimated travel required time is determined by the on-travel required time, and it is selectively increased according to the travel required time between the geographical location (6013) and the nearby representative location (6011),

(iv) obtain a required travel time 6014 between a commuting destination 6001 and a geographic location 6015 by means of a navigation service;

An example of a threshold is 1000 meters or 1 minute of travel,

In some embodiments, we find at least one neighborhood representative within a threshold of the commuting destination or geographic location, set the travel duration to be the minimum travel latency between the commuting destination and the geographic location that crosses all neighborhood representations,

In some aspects, we pre-calculate Voronoi cells for data about the nearest neighborhood (e.g., representative of distance or travel time), and determine the representative of the neighborhood while processing the request. Use data about this nearest neighborhood.

(b) one or more, but pre-determined limits, using the estimated travel time: the pre-determined limit will affect the number of clusters, the display of which will be determined later in step (d), within select clusters of listings in , the set based on at least one of, but not limited to: a number of clusters for which an indication must be included in the request, and a degradation in travel quality caused by the expected travel duration of step (a). , improvement in travel quality caused by the travel time required in step (c), consumption of information sources related to determining the required travel time in step (a), or consumption of information sources related to determining the required travel time in step (c). Consumption, e.g., with a predetermined limit of 1000 (e.g., selecting the highest-scoring clusters; e.g., selecting lists with the desired feature; clustering the selected lists and scoring these clusters); );

(c) determining travel times between commuting destinations and geographic locations listed in the selected clusters; For example, using the method of calculating travel mentioned in previous studies or disclosures of calculating the shortest path;

(d) using travel time (e.g. update the clusters and score using travel time; select clusters, e.g. a number of highest scoring clusters, this number at most, e.g. the 20 highest scores Determining the representation of selected clusters, such as clusters, within predetermined limits.

[065] In some aspects, the method also adjusts the travel around the representative location. The adjustment prevents unnatural shapes near the place where the pre-calculated travel merges with the increased travel. For example, we take a portion of a pre-calculated travel departing from a representative location near the departure location, and extend this partial travel to the travel between the locations included between the departure location and the partial travel (this location is need not be representative). Additional information on coordination can be found in the previous study WO2021222046.

[066] In some aspects, the method computes an overview using a two-step approach. In some embodiments, the method is a “first order list” ￡' ₁ , ￡' ₂ , . . . , ￡ _k' is calculated as above. However, instead of the required travel time, the estimated travel required time is used, and a narrow range of, for example, 1 minute is used. Since the range is narrow, the value of k' will often be large. Then, the method returns the lists ￡' ₁ , ￡' ₂ ,... , determines the required travel times of ￡'_k' . Then, this method can be used over a wider range, eg 15 minutes, but lists ￡' ₁ , ￡' ₂ , . . . , ￡'_k' is shuffled or organized, and the “first priority” lists ￡ ₁ , ￡ ₂ , ￡ ₃ ,… using the travel time required. , calculate ￡ _k . For example, this method can be used to list the lists £ ₁ , £ ₂ , £ ₃ ,… , computes ￡ _k , where the computation consists of the lists ￡' ₁ , ￡' ₂ ,... , starting from ￡ _k' (above, we gave an embodiment in which the calculation starts from list L). In some aspects, the method uses the expected travel time to calculate a geographic representative. Within the scope and spirit of the embodiments, many other methods of computing an overview using a two-step approach will be apparent to those skilled in the art.

[067] In some aspects, this method uses a two-step approach to our methods. In some aspects, the method uses a two-step approach to compute indications for at least two alternatives, where set A is determined using expected travel times and indications are determined using expected travel times. In some aspects, this method uses the two-step approach of our method. In some aspects, the method uses a two-step approach to compute an indication of at least two alternatives, where set A is determined using expected travel times and an indication is determined using the expected travel times. In some aspects, the method uses the estimated travel time to calculate an isochronous line and then calculates an indication. Within the scope and spirit of the embodiments, many other methods using the two-step approach of our method will be apparent to those skilled in the art.

1.3.5 Modified steps

[068] In some aspects, the sub-module performs modified steps compared to the steps described in the disclosure. For example, a submodule performs these steps: in a different order, partially concurrently, or combining or omitting certain steps. In some embodiments, the two steps of grouping or clustering are combined into one step. For example, we augment feature vectors. We take the list's feature vectors and append them: the features and values that contain the travel time between the list's geographic locations and at least one commuting destination, and the appropriate interesting features and values. The feature vectors expanded in this way become clustered. In some embodiments, clusters are determined using any clustering method used in the disclosure. In some aspects, clusters meet additional requirements. For example, we limit the extent of a certain cluster along the axis of added features, for example to 15 minutes at most. In some aspects, pre-computed clustering is used to accelerate clustering in the request processing phase. For example, using a clustering algorithm that starts running on precomputed clusters. In some aspects, the clustering step does not use similarity. In some aspects, a step finds clusters of at least one point. In some aspects, using the information contained in the request, as described in the disclosure, the step locates at least one point, such as: filtering constraints, or desired features. Within the scope and spirit of the embodiments, many other ways of carrying out the modified steps will be apparent to those skilled in the art.

2 Common cases

[069] We use the word travel in a broad sense consistent with what experts in the field understand. This word has meanings that include moving objects or data. Here are some examples of travel explanations: (1) “Hey buddy, you go one block north and then a little to the side.” (2) “Five dollars”. Length of travel is a numerical value that experts in the field can associate with travel, such as the monetary cost of travel; distance in meters, fuel consumption; Specific characteristics or attributes of travel descriptions, such as: number of transfers or walking distance. As another example, we can use the word travel time when we mean the length of a travel which means time. As another example, we can use the word travel given time when we mean the distance of a travel which means time. In some embodiments, the distance of the travel is derived using both endpoints of the travel, including: properties, commuting destinations, or their characteristics or values. For example, the length of the travel is derived using a weighted sum, i.e. the sum of two values: (1) the distance in meters from the exit of the company to the entrance of the building, (2) the number of floors of the property in the building. The value of the characteristic to represent. In some aspects, the distance of the travel is used on request. For example, a request includes an arbitrary conversion between two values. For example, the length of travel is the product of fuel consumption multiplied by a unit converted into an amount of money, and accordingly, the optimal target of fuel consumption is converted into a final target of monetary cost. In some embodiments, an arbitrary predetermined conversion to the length of the travel is used. In some embodiments, the use of two or more optimization goals combined into one optimization goal over the length of travel, eg using a mathematical formula such as a weighted sum. In some aspects, optimization goals are incorporated into multi-objective optimization based on multifactorial costs. For example, in order to reduce the monetary cost of travel, it is a method of searching for real estate that minimizes the travel time, and it is possible by filtering restrictions. The length of the travel is itself a description of the travel. Description of travel: It may not include the length of all travels, it may include only the length of one travel, or it may include other data as well. The disclosure teaches how to calculate the description of the travel, eg any method of calculating the travel described in the disclosure, or how to calculate the travel in the previous work mentioned in the disclosure, e.g. There is Dijikstra's Algorithm.

[070] We use the term transport system in a broad sense as understood by experts in the field. Some aspects include: systems of roads and vehicles; transportation systems including buses and subways; walkway system; airports, planes and routes; By boat or by sea. Transport systems do not necessarily physically move objects. The method of disclosure only needs to be able to determine the definition of travel between components of a transport system. Thus, a transport system that travels data is an example of a computer network that includes the following transport elements: wires/lines (similar to roads), hubs/switches (similar to stops/turns). The combination of all transportation systems that enable travel between them is a transportation system. Within the scope and spirit of the embodiments, other examples of transportation systems will be apparent to those skilled in the art.

[071] In some aspects, our invention is of interest other than the real estate aspect of searching and comparing. In some aspects, our method uses the similarity between the job posting and the travel time between the job site and the house to provide job postings. For aspects of search and comparison, reference may be made to the previous study WO2021222046.

[072] In general, the method of disclosure is an arbitrary site included in the transportation system (in the previous section, the site was named as real estate) and a random place (in the previous section, the location was named as a commuting destination). , and this method determines the display of at least one point by using the definition of travel between at least one point and one or more locations (the travel time used in the previous part has been generalized to the definition of travel). . A branch is an arbitrary place. This can be any property, for example an apartment, a rented room, a house with a garden, a ranch or a hotel. It can also be a place where people work, a restaurant, a store, or any other branch. A location is also an arbitrary place. Locations include schools, grandparents' homes, weekend golf courses, favorite restaurants, hospitals, and places of worship. It could also be where people live. In some aspects, a point of interest is interpreted as a site. In some aspects, the object of interest is interpreted as a place.

[073] An aspect is a method of searching or comparing using a description of at least one travel in a transportation system between at least one point and at least one location, the method including: (a) at least one receive requests including the location of; (b) respond to requests with the results of searches and comparisons obtained using at least one travel description; In some aspects, the result of a search or comparison is an aspect of an indication of at least one point.

[074] In some aspects, the methods of the disclosure perform variations on previously described functions or steps. In some embodiments, certain functions or procedures are performed partially or concurrently, in a different order, or certain functions or procedures are combined or omitted. For example, a method performs serving but not collection or indexing. In other instances, the indexing module 1003 does not create an inverted index 1005 or clustering 1006. In other instances, the request does not include specification of at least one commuting destination. In some aspects, the method performs clustering or scoring without using travel time between the property and travel time between the at least one or more commuting destinations. In some embodiments, the plurality of listings is determined using one or more of the following, but not both: (i) the similarity of the listings, or (ii) the relationship between at least one commuting destination at a geographic location in the listing. travel time. Within the scope and spirit of the embodiments, many other ways of performing variations of functions or steps will be apparent to those skilled in the art.

[075] The appearance of the invention may have a form of a hardware aspect, a software aspect, or a combination of the two. For example, any flow chart block may be served out of order, partially, concurrently, or from cache, depending on function or optimization. The shape of an invention can take the form of a continuous system or a parallel/distributed system in which each component embodies a certain shape while communicating with other components, possibly redundantly, for example, through some kind of network. there is. The present invention is not described based on any particular programming language. A computer program containing instructions for performing the steps of the outlines of the present invention may be written in any programming language. Examples include C++, Java, and JavaScript. Any program can run on any hardware platform. For example, a central processing unit (CPU) or a graphics processing unit (GPU) and related memory or storage devices may be mentioned. The program may implement aspects of the invention in one or more devices, but is not limited to: a smartphone operating Android or iOS operating system, or Firefox, Chrome, Internet Explorer. A web browser such as Safari.

3 way

[076] Aspects of the invention include the following methods.

1. A method for determining the marks of a plurality of points included in a transportation system using similarity and length of travel;

Features of the method include:

(a) receiving a request containing at least one location included in the transportation system;

(b) determining at least two isochronous points included in the plurality of points;

where the length of travel in the transport system between each isochronous point and the at least one location is included in the range;

(c) Determining the indication using steps including one of the following:

i. determining a plurality of similar points included in the at least two isochronous points, followed by determining the representation of the plurality of similar points; or

ii. selecting at least one first point dissimilar to the at least one second point, both of which are included in the at least two isochronous points, then selecting the at least one first point and the at least one first point. determining the indication of two points; also

(d) Responding to the request with the indication.

2. A method of determining an overview of a plurality of points included in a transport system using the number and length of travel;

Features of the method include:

(a) receiving a request containing at least one location included in the transportation system;

(b) Calculating an array of two or more points included in the plurality of points, wherein,

i. At a first point and a second point included in the arrangement, the length of travel in the transport system between the first point and the at least one location is equal to the distance between the second point and the at least one location in the transport system. at least a certain range away from the length of the travel, and

ii. If the length of travel in the transportation system between a fourth point and the at least one location is around the length of the travel in the transportation system between a third point and the at least one location, then the The number associated with the third point is at most the number associated with the fourth point included in the plurality of points;

(c) determining the overview including an indication of the arrangement; also

(d) Responding to the request with the overview.

3. a method for determining an indication of at least two alternatives included in a plurality of objects of interest included in the transportation system;

Features of the method include:

(a) receiving a request including a point included in the transportation system;

(b) determining the at least two alternatives;

The length of travel in the transportation system between each alternative and the point is at most the smallest distance plus the threshold;

(c) determining the indication for the at least two alternatives to be non-unitary and not descriptive of travel; also

(d) Responding to the request with the indication.

4. a method for determining the indication of at least two points included in the transport system using the length of travel and the length of expected travel;

Features of the method include:

(a) receiving a request containing at least one location included in the transportation system;

(b) determining the lengths of at least two expected travels;

wherein the at least two expected lengths of travel include a length of expected travel in the transportation system between the at least one location and each point included in the at least two points;

(c) Selecting one or more points included in the at least two points using the lengths of the at least two expected travels;

Here, the number of said one or more points is at most a predetermined limit;

(d) determining the length of at least one travel;

wherein the length of the at least one travel includes the length of the travel in the transportation system between the at least one location and each point included in the one or more points;

(e) determining the indication of the one or more points using the length of the at least one travel; also

(f) Responding to the request with the indication.

4 Computer Systems and Devices

[077] One aspect of the present invention is a computer system (eg Figure 1). A computer system has a hardware aspect, a software aspect, or a combination of both. A computer system includes at least one processor, for example a CPU or a GPU. A computer system includes a non-transitory computer-readable storage medium storing one or more programs executable by at least one processor. Aspects of non-transitory computer-readable storage media are well known in the art and require no elucidation here. In order to perform at least one step of the method described in the disclosure, one or more programs include instructions to be executed in at least one processor. In some aspects, the instructions are expressed in any computer language, for example C++, Java or Java Script. Each method creates a computer system. Any computer system can be viewed as a general-purpose computer specifically programmed to perform the specific methods described in the disclosure. Thus, in practice, this computer system is a special purpose computer programmed to perform the specific steps of the method according to instructions for software (one or more programs) encoding the method. Within the scope and spirit of the aspects, many other aspects of computer systems will be apparent to those skilled in the art.

[078] One of the aspects of the invention is an apparatus, also referred to as a device. The cases are presented in Figures 3, 4 and 5. Within the scope and spirit of the embodiments, it will be apparent to those skilled in the art that the presentation of the device in our figures may be modified (eg, rearrangement, adjustment of dimensions, change of color, addition or removal of shape, part). A device has a physical aspect, such as a smartphone app or a web page. The device is a “receiver”, e.g. the user interface of a smartphone app, e.g. when a user enters a work address into a search box, or taps a map displayed on a smartphone app to specify a place of work, the place of work It receives requests from the user, such as talking to the voice recognition engine to describe the job and providing GPS readings that code the location of the workplace. In some aspects, any information included in the request described in the disclosure is received. The device then generates an indication using the method described in the disclosure. In some embodiments, this generation is performed by executing one or more suitable programs on at least one processor. The device then responds to the user by presenting an indication via the “sender”, e.g.: display of a smartphone app (e.g., display 3005 as depicted in FIG. 3), voice synthesizer (e.g., to the user talking), augmented reality lenses included in eyeglasses worn on a user's head, 3D projectors of Jedi Knights, etc. It will be apparent to those skilled in the art that the present invention does not apply only to a particular device, receiver or transmitter. • Many other aspects of computer services will be apparent to those skilled in the art.

Within the scope and spirit of the embodiments, many other aspects of the apparatus will be apparent to those skilled in the art.

5 conclusions

[079] It will be appreciated by those skilled in the art that, within the scope and spirit of the embodiments, many modifications are possible, and alternatives that are essentially the same. Also within the scope and spirit of the aspects, certain circumstances may apply to the teachings of the invention. Accordingly, even if the invention has been described based on closed aspects, the invention will not be limited only to these aspects. Rather, the invention will include all aspects falling within the scope of the appended claims.

[080] Each claimed method does not involve a "mental process" (none of the claimed steps were performed in the human mind). All claimed methods are automated. Section 4 describes examples of automation. The scope of each claimed method excludes any aspect that would be unpatentable in the particular jurisdiction in which this patent application was filed during the PCT country/region phase. For example, an application for a patent in Canada implicitly states that each claimed method is limited to only aspects that are eligible for a patent in Canada. Each particular jurisdiction excludes aspects specific to that jurisdiction (other jurisdictions may exclude other sets of aspects).

[081] In some embodiments, any claimed method is performed on a computer system (computer-implemented) and achieves a purpose on a device (eg, for purposes of searching or comparing or determining a display). Section 4 explains the case. In any aspect, any claimed method is limited to that which falls within the scope of a "method of manufacture" within the meaning of New Zealand's monopoly law, as will be clear to experts in the field. In any aspect, any method of assertion is limited to aspects falling within the scope of a "technical character" within the meaning of the European Patent Convention, as will be clear to experts in the field.

[082] Any previous work cited in the invention is considered common knowledge in this field; Anyone in this field has this knowledge.

[083] Antecedent basis is sometimes traced in a claim box: a term in a claim may be used in a later term.

[084] We have included a glossary of selected terms appearing in the claims and example references to specific circumstances. These references are not exhaustive, and other references exist. The arrangement of terms in the table was intended to be written in the order in which these terms appeared in the claim.

term (phrase) example reference indication [038] section branch (site) [072] section plurality of sites (lists, clusters of lists, clusters, groups) Sections [034], [035], [068], [074] transportation system [070] section description of travel [069] section similarity Section [004] request Sections [031], [035], [036], [037], [078] location [072] section at least one place (at least one commute destination) [031] section isochrone site Sections [034], [068] length of travel [069] section range Sections [034], [068], [042] similar sites (similar lists, clusters) Sections [035], [068] selection Sections [037], [040], [064], [068] not similar Sections [040], [049] First site, second site (first site, second site) Sections [040], [049] overview [041] section array (sequence) [042] section neighborhood [042] section at least two alternatives [053] section point of interest Sections [020], [062] threshold [053] section not a description of travel [055] section non-singular [056] section Estimated length of travel (estimated travel duration) Sections [033], [064] predetermined bound [064] section

Claims (4)

A method for determining the indication of a plurality of points included in a transportation system using similarity and length of travel;
Features of the method include:
(a) receiving a request containing at least one location included in the transportation system;
(b) determining at least two isochronous points included in the plurality of points;
where the length of travel in the transport system between each isochronous point and the at least one location is included in the range;
(c) determining the indication using steps comprising one of the following:
i. determining a plurality of similar points included in the at least two isochronous points, followed by determining the representation of the plurality of similar points; or
ii. selecting at least one first point dissimilar to the at least one second point, both of which are included in the at least two isochronous points, then selecting the at least one first point and the at least one first point. determining the indication of two points; also
(d) responding to the request with the indication. A method for determining an overview of a plurality of points included in a transport system using the number and length of travel;
Features of the method include:
(a) receiving a request containing at least one location included in the transportation system;
(b) calculating an array of two or more points included in the plurality of points, where:
i. At a first point and a second point included in the arrangement, the length of travel in the transportation system between the first point and the at least one location is equal to the distance between the second point and the at least one location in the transportation system. at least a certain range away from the length of the travel, and
ii. If the length of travel in the transportation system between a fourth point and the at least one location is around the length of travel in the transportation system between a third point and the at least one location, then the The number associated with the third point is at most the number associated with the fourth point included in the plurality of points;
(c) determining the overview including an indication of the arrangement; also
(d) responding to the request with the overview. a method for determining an indication of at least two alternatives included in a plurality of objects of interest included in a transportation system;
Features of the method include:
(a) receiving a request including a point included in the transportation system;
(b) determining the at least two alternatives;
The length of travel in the transportation system between each alternative and the point is at most the smallest distance plus the threshold;
(c) determining the indication for the at least two alternatives to be non-unitary and not descriptive of travel; also
(d) responding to the request with the indication. a method for determining an indication of at least two points included in a transportation system using the length of travel and the expected length of travel;
Features of the method include:
(a) receiving a request containing at least one location included in the transportation system;
(b) determining the lengths of at least two expected travels;
wherein the at least two expected lengths of travel include a length of expected travel in the transportation system between the at least one location and each point included in the at least two points;
(c) selecting one or more points included in the at least two points using the lengths of the at least two expected travels;
Here, the number of said one or more points is at most a predetermined limit;
(d) determining the length of at least one travel;
wherein the length of the at least one travel includes the length of the travel in the transportation system between the at least one location and each point included in the one or more points;
(e) determining the indication of the one or more points using the length of the at least one travel; also
(f) responding to the request with the indication.